Toxic Takifugu Pardalis Eggs Found in Takifugu Niphobles Gut: Implications for TTX Accumulation in the Pufferfish

Toxicon 108 (2015) 141e146

Contents lists available at ScienceDirect

Toxicon

journal homepage: www.elsevier.com/locate/toxicon

Toxic Takifugu pardalis eggs found in Takifugu niphobles gut: Implications for TTX accumulation in the pufferﬁsh

* Shiro Itoi a, , Ao Kozaki a, Keitaro Komori a, Tadasuke Tsunashima a, Shunsuke Noguchi a, 1, Mitsuo Kawane b, Haruo Sugita a a Department of Marine Science and Resources, Nihon University, Fujisawa, Kanagawa 252-0880, Japan b Department of Sea-Farming, Aichi Fish Farming Institute, Tahara, Aichi 441-3618, Japan article info abstract

Article history: Pufferfish (Takifugu spp.) possess a potent neurotoxin, tetrodotoxin (TTX). TTX has been detected in Received 9 July 2015 various organisms including food animals of pufferfish, and TTX-producing bacteria have been isolated Received in revised form from these animals. TTX in marine pufferfish accumulates in the pufferfish via the food web starting with 30 September 2015 marine bacteria. However, such accumulation is unlikely to account for the amount of TTX in the puf- Accepted 14 October 2015 ferfish body because of the minute amounts of TTX produced by marine bacteria. Therefore, the tox- Available online 19 October 2015 ification process in pufferfish still remains unclear. In this article we report the presence of numerous Takifugu pardalis eggs in the intestinal contents of another pufferfish, Takifugu niphobles. The identity of Keywords: Congeneric eggs T. pardalis being determined by direct sequencing for mitochondrial DNA. LC-MS/MS analysis revealed fi Food chain that the peak detected in the egg samples corresponded to TTX. Toxi cation experiments in recirculating Pufferfish aquaria demonstrated that cultured Takifugu rubripes quickly became toxic upon being fed toxic (TTX- Takifugu niphobles containing) T. rubripes eggs. These results suggest that T. niphobles ingested the toxic eggs of another Takifugu pardalis pufferfish T. pardalis to toxify themselves more efficiently via a TTX loop consisting of TTX-bearing or- Tetrodotoxin ganisms at a higher trophic level in the food web. © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction pufferfish and the potential food organisms (Noguchi et al., 1987; Wu et al., 2005; Noguchi and Arakawa, 2008). In addition, TTX Tetrodotoxin (TTX) is known to be the substance of pufferfish has been detected in some free-living bacteria, including those in toxin and one type of potent neurotoxin specific to voltage-gated deep sea sediments (Simidu et al., 1987; Do et al., 1990), although it sodium channels of excitable membranes of muscle and nerve is not clear if these bacteria form part of the food chain leading to tissues (Colquhon et al., 1972; Narahashi, 2001; Noguchi et al., pufferfish. In any case, it appears plausible that the TTX in pufferfish 2006a). TTX was believed to occur only in pufferfish (Tetraodonti- is a result of accumulation through the food chain, which consists of dae) until 1960's, when it was detected in Californian newt Taricha several steps, starting with bacteria, as suggested by several reports torosa (Mosher et al., 1964). Subsequently, TTX (along with some (Noguchi et al., 2006a; Noguchi and Arakawa, 2008). These spec- analogs) was also detected from potential pufferfish food organ- ulations have actually been supported by several studies: non-toxic isms belonging to various disparate groups, including starfish (e.g., pufferfish have been produced when grown from hatching with a Astropecten spp.; Maruyama et al., 1984, 1985), gastropods (e.g., non-toxic diet, and furthermore, these cultured non-toxic puffer- Babylonia japonica; Noguchi et al., 1981), crustaceans (e.g., the fish have become toxic when administered orally with TTX (Matsui xanthid crab, Atergatis floridus; Noguchi et al., 1983), flatworms and et al., 1981, 1982; Noguchi et al., 2006b; Saito et al., 1984; Yamamori ribbonworms (e.g., Cephalothrix simula; Asakawa et al., 2013), apart et al., 2004; Honda et al., 2005). from several species of bacteria that are symbiotic with the TTX has been detected not only in pufferfish and their prey, but also in organisms ecologically unrelated to pufferfish, such as the Costa Rican frogs of the genus Atelopus (Kim et al., 1975) and some * Corresponding author. land planarians (Stokes et al., 2014), besides Californian newt E-mail address: [email protected] (S. Itoi). T. torosa (Mosher et al., 1964). It has also been suggested that TTX in 1 Present address: Kyoto Institute of Oceanic and Fishery Science, Miyazu, Kyoto the rough-skin newt, Taricha granulosa, is obtained endogenously 626-0052, Japan. http://dx.doi.org/10.1016/j.toxicon.2015.10.009 0041-0101/© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 142 S. Itoi et al. / Toxicon 108 (2015) 141e146

(Hanifin et al., 2002; Cardall et al., 2004). These findings suggest extraction and the remaining eggs were stored at 30 C until TTX that TTX accumulates in pufferfish by means other than through extraction. the classical food chain; since in vivo cultured TTX-producing bacteria are unable to produce enough quantities of TTX to ac- 2.2. DNA extraction and PCR amplification count for the amount of TTX in wild pufferfish (Miyazawa and Noguchi, 2001; Food Safety Commission of Japan, 2005; Wu et al., Total genomic DNA was extracted from each egg by the method 2005; Wang et al., 2008; Yang et al., 2010). This also indicates of Sezaki et al. (1999). The fragments of partial mitochondrial DNA that TTX-bearing pufferfish prey are abundant. In any case, the were amplified by PCR using two primer sets, 16S AR-L (forward: 50- origin of TTX and the acquisition process are both likely to vary CGCCT GTTTA TCAAA AACAT-30) and 16S BR-H (reverse: 50-CCGGT across different animal species, and it remains unclear exactly CTGAA CTCAG ATCAC GT-30) for the 16S rRNA gene (Palumbi et al., where pufferfish acquire the large quantities of TTX that they 1991), and TCytb-F1 (forward: 50-ACCTR TGGCG TGAAA AACCA possess. YCGTT GT-30) and TCytb-R1 (reverse: 50-CATYC GGTTT ACAAG Recently, our lab unexpectedly found numerous eggs among the ACCGR CGCTC TG-30) for the cytochrome b gene. Primers for cyto- intestinal contents of the pufferfish, Takifugu niphobles, and partial chrome b gene were designed based on the mitochondrial DNA mitochondrial DNA sequences from these eggs identified them to sequences from multiple species of the family Tetraodontidae. PCR be those of another pufferfish species, namely, Takifugu pardalis.We amplification was performed in a 20 ml reaction mixture containing have demonstrated the toxification process using cultured puffer- genomic DNA as a template, 1 unit ExTaq DNA polymerase (Takara fish Takifugu rubripes in this study by means of experimentally Bio, Shiga, Japan), 1.6 ml of 2.5 mM deoxynucleotide triphosphates reproducing the serendipitous finding of eggs in the pufferfish gut, (dNTP), 5 mlof5mM primers and 2 mlof10 ExTaq DNA polymerase thus indicating that pufferfish toxification manifests from the buffer (Takara Bio). The thermal cycling program for the PCR con- accumulated TTX at relatively higher trophic levels in the food sisted of an initial denaturation at 95 C for 1 min followed by 35 chain. cycles of denaturation at 95 C for 10 s, annealing at 55 C for 30 s and extension at 72 C for 45 s. 2. Materials and methods 2.3. Direct sequencing and phylogenetic analyses 2.1. The intestinal contents of wild pufferfish Prior to the direct sequencing of the amplified product, the DNA Wild specimens of the pufferfish, T. niphobles (body weight: fragment was purified by chloroform extraction, followed by 60.3 ± 25.7 g and 48.9 ± 21.2 g for females and males, respectively; polyethylene glycol (PEG) 8000 precipitation and ethanol precipi- detailed in Table 1) were collected from coastal waters in Nagai, tation. Sequencing was performed for both strands using a 3130xl Kanagawa, Japan (35120N, 139360E), on 13 March 2012 (water genetic analyzer (Applied Biosystems, Foster, CA, USA) and a BigDye temperature: 13.9 C; salinity: 31.4 practical salinity unit, psu), 08 Terminator v3.1 Cycle Sequencing Ready Reaction Kit (Applied March 2013 (16.0 C; 32.4 psu) and 14 March 2013 (14.2 C; 30.6 Biosystems). The concatenated nucleotide sequences of the 16S psu), 07 March 2014 (12.5 C; 32.3 psu) and 13 March 2014 (12.5 C; rRNA gene and cytochrome b gene of eggs were aligned using 31.8 psu), and 09 March 2015 (11.7 C; 34.3 psu), 16 March 2015 CLUSTAL W (Thompson et al., 1994) with those in the DDBJ/EMBL/ (12.3 C; 34.8 psu) and 19 March 2015 (14.6 C; 35.0 psu). The GenBank databases obtained using a BLAST search (Altschul et al., gonadosomatic index (GSI) was 2.2 ± 0.9 for females and 2.0 ± 1.4 1997). The alignment was then subjected to phylogenetic infer- for males (detailed in Table 1). The unexpected eggs that were ence by means of the maximum likelihood method using MEGA ver. found among the gut contents of the pufferfish were also collected. 6.0.6 (Tamura et al., 2013), with the corresponding concatenated Five eggs from each fish were immediately subjected to DNA sequences from the yellow-stripe toadfish, Torquigener brevipennis

Table 1 Characteristics of the Takifugu niphobles specimens used in this studya.

Sampling Sex No. of No. of egg-fed Intestinal egg content Toxicity of the intestinal eggs Standard length Body weight GSIb date specimen specimen (g) (MU) (mm) (g)

2012 13 Mar Female n ¼ 8n¼ 8 N/Dc N/D 119.4 ± 3.8 64.1 ± 9.4 2.6 ± 0.5 Male n ¼ 2n¼ 2 N/D N/D 111.0 ± 2.0 54.5 ± 1.5 1.5 ± 0.2 2013 08 Mar Female n ¼ 9n¼ 8 0.28 ± 0.25 2.63 ± 4.42 112.7 ± 18.3 60.3 ± 22.8 2.1 ± 0.8 Male n ¼ 13 n ¼ 7 0.05 ± 0.09 0.25 ± 0.74 108.4 ± 17.7 55.2 ± 22.0 1.9 ± 1.1 14 Mar Female n ¼ 6n¼ 6 0.72 ± 0.48 10.38 ± 12.94 122.2 ± 16.7 56.2 ± 24.5 2.0 ± 0.4 Male n ¼ 8n¼ 8 0.26 ± 0.36 4.10 ± 6.48 110.8 ± 11.1 55.4 ± 12.2 2.7 ± 0.8 2014 07 Mar Male n ¼ 2n¼ 2 0.44 ± 0.30 0.29 ± 0.17 128.0 ± 1.0 73.3 ± 3.6 2.6 ± 0.3 13 Mar Female n ¼ 10 n ¼ 7 0.38 ± 0.62 0.49 ± 0.66 121.7 ± 17.2 63.9 ± 23.8 2.4 ± 1.3 Male n ¼ 6n¼ 3 0.09 ± 0.12 0.06 ± 0.13 111.3 ± 12.5 50.0 ± 17.4 3.0 ± 2.1 2015 09 Mar Female n ¼ 13 n ¼ 0 N/Ad N/A 118.7 ± 12.8 33.0 ± 9.7 1.9 ± 0.5 Male n ¼ 10 n ¼ 3 0.05 ± 0.08 2.72 ± 7.04 115.6 ± 17.4 30.8 ± 15.4 1.3 ± 1.3 16 Mar Female n ¼ 11 n ¼ 8 0.38 ± 0.57 15.84 ± 30.37 164.6 ± 17.6 87.3 ± 20.9 2.5 ± 0.7 Male n ¼ 4n¼ 2 0.08 ± 0.08 1.63 ± 2.32 140.0 ± 26.0 59.0 ± 28.0 2.1 ± 1.1 19 Mar Female n ¼ 4n¼ 4 0.47 ± 0.37 3.30 ± 2.46 130.0 ± 18.1 43.5 ± 16.0 2.5 ± 1.6 Male n ¼ 4n¼ 0 N/A N/A 133.0 ± 22.5 45.5 ± 18.1 2.0 ± 1.2

a Data are represented by mean ± standard deviation. b GSI represents gonadosomatic index: gonad weight/body weight 100. c N/D, not determined. d N/A, not applicated. S. Itoi et al. / Toxicon 108 (2015) 141e146 143

(DNA database accession number AP009537) as the outgroup. The was analyzed by means of a Student's t-test and a one-way ANOVA nucleotide sequences of the partial 16S rRNA gene and cytochrome followed by the Tukey Honestly Signiﬁcant Difference (HSD) test. In b gene obtained from the eggs were submitted to the DDBJ/EMBL/ order to evaluate statistical support values for the linear regression, GenBank databases with accession numbers LC060064 and results of toxicity from toxiﬁcation experiments were log- LC060065, respectively. transformed and subjected to ANOVA.

2.4. Toxiﬁcation experiment 3. Results

Toxic eggs for the toxification experiments were obtained from 3.1. Eggs from the intestinal contents of the pufferfish T. niphobles the pufferfish T. rubripes, which had been collected at Ise Bay, Aichi, Japan, and maintained in the aquarium at the Aichi Sea Farming A number of eggs, with a diameter of approximately 1 mm, were Institute with flowing water (5.2 t/h). The non-toxic juveniles of detected in the gut contents of the pufferfish T. niphobles (Fig. S1) T. rubripes (body weight: approximately 3.5 g) were purchased from captured in March 2012e2015 (Table 1). The eggs were seen in the Marinetech (Aichi, Japan), and were maintained in recirculating 08 March 2013 sample in 8 and 7 specimens of 9 females and 13 aquariums at 20 C, and fed with commercial food pellets (non- males, respectively. The eggs were detected in all the specimens in toxic). In the toxification experiment, juvenile T. rubripes (n ¼ 52, the 14 March 2013 sample. The eggs were seen in 7 and 5 speci- body weight: 3.1e49.6 g, 19.0 ± 12.7 g) were fed with the toxic eggs. mens of 10 females and 8 males, respectively, in 2014, and in 12 and After more than two days of feeding with toxic eggs, liver, skin, 5 specimens of 28 females and 18 males, respectively, in 2015 muscle and other organ samples of egg-fed T. rubripes and com- (Table 1). In a few fish, the eggs were yet to be digested in the mercial pellet-fed control specimens were subjected to the TTX anterior parts of the intestine that interestingly, hatched after in- extraction process followed by LC-MS/MS analysis. In advance before cubation at 20 C in artificial seawater for 9 days (March 2013). the toxification experiment, preliminary experiment was carried Direct sequencing and phylogenetic analysis of egg samples out: after two, four and nine days of feeding with toxic eggs, egg-fed from the intestinal contents of T. niphobles demonstrated that the T. rubripes specimens (n ¼ 27, body weight: 14.8 ± 3.6 g) were also sequence of 16S rRNA gene and cytochrome b gene were highly subjected to the TTX extraction followed by LC-MS/MS analysis. similar to and clustered with the corresponding sequences of T. pardalis (Fig. 1). 2.5. LC-MS/MS analysis

Following Tsuruda et al. (2002) with some modifications, the 3.2. Toxicity of the eggs in the intestinal contents of T. niphobles and extract with 0.1% acetic acid was filtered through a 0.45-mm filter tissue-distribution of TTX in T. niphobles fed T. pardalis-eggs membrane (Dismic-13CP, Advantec, Japan), and TTX in the extract was analyzed using High Performance Liquid Chromatography Eggs were collected from the intestines of T. niphobles, subjected (HPLC) equipment ACQUITY system (Alliance 2695), coupled to a to TTX extraction, and analyzed using LC-MS/MS. Mass chromato- Quattro Premier XE mass spectrometer from Waters (Milford, MA, gram of the LC-MS/MS was obtained under the MRM mode, and the USA), based on the method of Shinno et al. (2007) with some MRM patterns of the egg samples were found to be identical to modifications. Chromatographic separation and detection of TTX those of the TTX standards (Fig. 2). The calibration curve generated was achieved with a Waters Atlantis HILIC silica column with 1e100 ng/ml of TTX standard shows good linearity and (2.1 150 mm, 5 mm), equipped with a Waters Atlantis HILIC silica guard column (2.1 10 mm, 5 mm) and column oven at 40 C. The HPLC was operated with 0.1% formic acid (Merck, Darmstadt, Ger- 100 Eggs many) and acetonitrile (Merck) as eluents. The gradient protocol 100 Takifugu pardalis (AP009528) used to elute the toxins was 95% acetonitrile, mobile phase in the 67 Takifugu chrysops (AP009525) beginning, decreasing to 40% acetonitrile after 0.1 min, then kept 67 Takifugu xanthopterus (AP009533) for 7.9 min and back to 95% acetonitrile and finally kept 95% Takifugu ocellatus (AP009536) acetonitrile for 2.0 min before the next injection. Flow rate was 0.3 ml/min and injection volume was 5 ml. Takifugu niphobles (AP009526) 88 The Quattro Premier XE mass spectrometer was operated with 79 Takifugu snyderi (AP009531) the following optimized source-dependent parameters (ESI 100 Takifugu stictonotus (AP009530) source): capillary potential 0.5 kV, cone voltage 42 V, desolvation Takifugu porphyreus (AP009529) temperature 400 C, desolvation gas flow 600 l/h N2, cone gas flow 63 Takifugu rubripes (AP006045) 50 l/h N2, source temperature 120 C, collision energy 38 V. 100 Takifugu chinensis (AP009534) The mass spectrometer was operated in the multiple reaction monitoring (MRM) mode, with detection in the positive mode, Takifugu oblongus (AP009535) analyzing two product ions at m/z 162 for quantification of TTX and Torquigener brevipennis (AP009537) m/z 302 for confirmation of the compound from the precursor ion 0.02 at m/z 320. The ionized molecules were monitored through a MassLynx NT operation system. TTX was quantified using their Fig. 1. Phylogenetic tree inferred using the maximum likelihood method based on the peak areas to calculate amounts and using the calibration curve nucleotide sequence of the concatenation of the complete cytochrome b gene (1137 bp) and the partial 16S rRNA gene (572 bp), obtained from the eggs removed obtained from 1 to 100 ng/ml TTX standard (Wako Pure Chemicals, from the intestinal tract of Takifugu niphobles and those of other Takifugu pufferfish Osaka, Japan). One mouse unit (MU) is equivalent to 0.22 mg of TTX, species. Numbers at branches denote the bootstrap percentages out of 1000 replicates. based on the specific toxicity of TTX. The concatenated cytochrome b and 16S rRNA genes in Torquigener brevipennis (AP009537) was used as the outgroup. Only bootstrap values exceeding 50% are pre- sented. The scale at the bottom of the tree indicates the number of substitutions per 2.6. Statistical analysis nucleotide site. “Eggs” represent the corresponding concatenation of the two gene sequences of the eggs taken from the intestinal tract of T. niphobles. The accession The statistical significance of differences in the amount of toxin numbers for reference sequences are shown in parentheses. 144 S. Itoi et al. / Toxicon 108 (2015) 141e146 precision (r2 ¼ 0.99991). To estimate the effect of the matrix, LC- The whole-body toxicity of the T. niphobles specimens fed with MS/MS analysis was carried out using the samples, where the T. pardalis eggs was 2803 ± 10,361 MU (617 ± 2279 mg) for females TTX standard was added into various tissue extracts from the non- and 1901 ± 1856 MU (418 ± 408 mg) for males (Table S1). The tissue- toxic T. rubripes following the procedure mentioned above. TTX was specific toxicity in females was significantly higher in the skin and recovered from the samples with >100-fold dilution except for ovaries than in the liver (P < 0.01), whereas in the males it was gallbladder (>1000-fold dilutions) (Fig. S2). Therefore, quantifica- significantly higher in the skin than in other tissues (P < 0.01; tion of TTX was carried out using the data for samples with >100- Fig. S3). Similar patterns of toxin distribution were observed in the fold dilution. The concentrations of TTX in the eggs from intestinal samples from each year (2013e2015; Fig. S3). contents of the samples in 2012 were calculated to be 6.58 ± 0.88 mg/g, which correspond to 29.9 ± 4.0 MU/g; however, 3.3. Intoxication of the pufferfish T. rubripes fed toxic eggs the contents of the eggs were not quantified because all eggs were not collected (Table 1). The concentration of TTX in the eggs In preliminary toxification experiment, no significant difference collected from the gut of the 2013 specimens of T. niphobles was in the toxicity of whole body was observed among the T. rubripes estimated to be 2.22 ± 4.51 mg/g (0.97 ± 1.83 mg/intestinal contents specimens fed with TTX-containing eggs after being fed toxic eggs per fish), which corresponds to 10.1 ± 20.5 MU/g (4.4 ± 8.3 MU/ for two days (98 ± 109 ng/g), four days (139 ± 296 ng/g) and nine intestinal contents per fish). Similar patterns of the concentration days (154 ± 270 ng/g), suggesting that the pufferfish keeps the of TTX in the eggs from intestinal contents of the T. niphobles were toxin level for a while (at least nine days; Fig. S4). Therefore, tox- observed in the samples from each year (2013e2015; Table S1). ification experiments were performed after being fed toxic eggs for Additionally, the TTX amount in the intestinal materials from egg- more than two days. fed T. niphobles specimens (4139 ± 6023 ng) were significantly Non-toxic T. rubripes (n ¼ 31, body weight: 3.1e49.6 g, higher than those from non-egg-fed counterparts (216 ± 374 ng) in 21.9 ± 12.8 g) became toxic after being fed toxic eggs for more than 2015 (Table S2). two days, although some specimens (n ¼ 21, body weight: 3.1e42.0 g, 14.7 ± 11.1 g) continued to remain non-toxic. The MRM patterns of the toxified tissues from the intestine, liver and skin 150000 were identical to those from the TTX extracted from the toxic eggs and the TTX standards (Fig. 2). The amount of toxin detected from several organs of toxified specimens was as follows: skin, 100000 47.5 ± 38.1 mg/g; liver, 31.8 ± 31.6 mg/g; and intestine, 19.9 ± 29.0 mg/ g(Fig. S5). The amount of TTX in toxified fish was dependent on its body weight in an exponential relationship (y ¼ 0.9566e0.1307x, R2 ¼ 0.6402; Fig. 3) with high statistical support values for the 50000 (A) Standard (25 ng/ml) linear regression (P ¼ 5.46 10 7, F statistics ¼ 44.49). On the other hand, none of the control fish fed with commercial feed were toxified (control; n ¼ 38, body weight: 3.1e57.9 g, 15.6 ± 15.6 g). 0 Furthermore, the fish that did not ingest the toxic eggs even when given were not toxified either. 50000 (B) Intestinal eggs 4. Discussion

fi 0 Non-toxic T. rubripes were rapidly toxi ed after feeding on eggs containing TTX. Similar profiles of TTX transfer were observed in 50000 (C) Eggs of T. rubripes the cultured T. rubripes when TTX was administrated intramuscularly (Ikeda et al., 2009). All species of the genus Takifugu accu- Signal intensity mulate TTX in the ovaries, with the ovaries of T. pardalis showing the maximum toxicity of more than 1000 MU/g tissue, which cor- 0 responds to approximately 220 mg TTX/g tissue (classified as “strongly toxic”, Noguchi et al., 2006a). These suggest that 50000 (D) Skin of intoxicated T. rubripes

10000

y = 0.9566e0.1307x 0 1000 R2 = 0.6402

50000 (E) Liver of non-toxic T. rubripes 100

10 0 1

0246810 Total amount of TTX (MU) Retention time (min) 0.1 nd Fig. 2. Typical mass chromatograms of the LC-MS/MS obtained under MRM mode (m/z 0 102030405060 320 > 302). MRM patterns of 25 ng/ml TTX standard (A), the extract from eggs found in Body weight (g) the intestinal tract of T. niphobles (B), the extract from the toxic eggs of T. rubripes used for the intoxication experiment (C), the extract from the skin of T. rubripes after the Fig. 3. The relationship between the total amount of TTX in a specimen and the body intoxication experiment (D), and the extract from the liver of non-toxic T. rubripes weight of T. rubripes after the experiment with toxic eggs. The amount of TTX was cultured with only commercial feed (E). calculated from the skin, liver, intestine and other tissues. nd: not detected. S. Itoi et al. / Toxicon 108 (2015) 141e146 145

TTX Pufferfish

TTX TTX TTX Food animals for pufferfish

TTX TTX TTX Marine bacteria

Fig. 4. TTX accumulation process in pufferfish on the TTX loop. Upward arrows represent general TTX accumulation process in pufferfish via food chain consisting of several steps starting with marine bacteria (Noguchi et al., 2006a). A circular arrow represents novel TTX accumulation process in Takifugu niphobles (colored fish) at the spawning season of Takifugu pardalis (grey color fish) observed in this study.

T. niphobles effectively ingest TTX along with the nutrients in the predators, and the other as an attractant for TTX-bearing organ- toxic eggs of other pufferfish (T. pardalis in this study): our present isms. The latter relationship would result in the accumulation of study also demonstrated that the TTX amount in the intestinal TTX at higher trophic levels in the food chain (Fig. 4). Additionally, materials from toxic egg-fed T. niphobles was significantly higher we speculate that the biomass of TTX-bearing organisms might be than that from non-egg-fed specimens. We noted this ingestion of controlled (limited) by the amount of TTX in the marine environ- eggs in four consecutive years (2012e2015), leading us to believe ment, because the toxin content of TTX-bearing organisms could that this is part of the normal diet of these fish; indeed, it is possible still be limited by available resources (Williams, 2010). that other Takifugu pufferfish species also adopt a similar strategy In this study, the amount of TTX in the toxified T. rubripes was for toxification. The findings in this study perhaps remove one of found to be a function of the size (surface area) of young fish, several objections to postulated pathways of TTX accumulation in suggesting that the largest young fish were also the ones with the pufferfish via the marine food web (Miyazawa and Noguchi, 2001; greatest amount of TTX. On the other hand, some specimens of Food Safety Commission of Japan, 2005; Wu et al., 2005; Wang T. niphobles, the smallest species in the genus, contained a large et al., 2008; Yang et al., 2010). In TTX-bearing newts in the genus amount of TTX (>100,000 MU in one individual; Tani, 1945). It has Taricha, adults prey on both conspecific eggs and larvae (Kats et al., also been suggested that the maximum toxin content of the puf- 1992; Elliott et al., 1993), suggesting that these cannibalistic egg/ ferfish might be affected by the various physiological factors larva consumptions may also play a role in their toxification. including sexual development (Ikeda et al., 2010; Itoi et al., 2012). In Interestingly, the eggs removed from the intestines of the toxification experiments, no TTX was detected from several T. niphobles hatched after incubation for 9 days at 20 C. This sug- specimens, suggesting that these specimens did not feed upon the gests that the pufferfish T. niphobles ingested the T. pardalis eggs toxic eggs in the aquaria. Differences in the feeding behavior among within one week of spawning, implying that they ingest highly pufferfish should affect the distribution of toxicity in wild pufferfish concentrated TTX via the eggs, because the toxicity of the eggs is specimens; in fact, the toxicity of pufferfish does show remarkable known to be highest immediately after spawning (Itoi et al., 2014). individual and regional variation (Miyazawa and Noguchi, 2001). Nevertheless, the TTX content in the intestinal eggs from In conclusion, predation on toxic eggs of related species should T. niphobles were relatively low, suggesting that the TTX was enable Takifugu pufferfish to accumulate TTX more efficiently than immediately absorbed in intestine and transfer to other tissues if dependent only on the classical food chain. Our results also such as liver, skin and ovary: the speculation was supported by the suggest that concentrated TTX is pooled in the “TTX loop” among results of toxification experiment using T. rubripes in the present the TTX-bearing organisms, which are the predators at higher study. The spawning period of T. niphobles begins in May, sug- trophic levels in the food web. gesting that the pufferfish obtain TTX and nutrients required for their forthcoming spawning, when they feed on T. pardalis eggs (in the spring). Acknowledgments Toxic organisms such as pufferfish use TTX in two ways: as a defense mechanism against predators by releasing it from the toxin We express sincere thanks to the staff of Kanagawa Prefectural gland in the skin (Kodama et al., 1985), and protecting the larvae by Marine Science High School for help in collecting samples. We also fi accumulating it in the ovaries (Hani n et al., 2003; Itoi et al., 2014), express sincere thanks to K.I., N.T., R.M., N.Y., S.T., S.D. and T.O. of and in predation on non-toxic organisms, such as the blue-ringed Nihon University College of Bioresource Sciences for help in col- octopus Hapalochlaena maculosa (Sheumack et al., 1978) and in lecting samples and rearing the pufferfish. This study was sup- fl polyclad atworms (Ritson-Williams et al., 2006). On the other ported in part by Research Grants for 2010 and 2013 from the Nihon hand, presentation of TTX against TTX-bearing predators appears to University College of Bioresource Sciences (S.I.), Grant-in-Aid for induce the opposite effect, where TTX has been reported as an Young Scientists (A) from Japan Society for the Promotion of Sci- attractant in trumpet snails (Hwang et al., 2004) and T. rubripes ence (JSPS) (23688023, S.I.), Grant-in-Aid for challenging Explor- (Saito et al., 1997; Okita et al., 2013). Thus, concentrated TTX can atory Research from JSPS (26660177, S.I.), and Grant-in-Aid for serve a dual function: one as a chemical defense against non-toxic Scientific Research (B) from JSPS (15H04552, S.I.). 146 S. Itoi et al. / Toxicon 108 (2015) 141e146

Conflict of interest tetrodotoxin in the starfish Astropecten latespinosus. Experimentia 40, 1395e1396. Maruyama, J., Noguchi, T., Narita, H., Jeon, J.K., Otsuka, M., Hashimoto, K., 1985. The authors declare that there are no conflicts of interest. Occurrence of tetrodotoxin in a starfish, Astropecten scoparius. Agric. Biol. Chem. 49, 3069e3070. Matsui, T., Hamada, S., Konosu, S., 1981. Difference in accumulation of puffer fish Ethical statement toxin and crystalline tetrodotoxin in the puffer fish, Fugu rubripes rubripes. Bull. Jpn. Soc. Sci. Fish. 47, 535e537. The authors declare that this manuscript complies with the Matsui, T., Sato, H., Hamada, S., Shimizu, C., 1982. Comparison of toxicity of the fi Elsevier Ethical Guidelines for Journal Publication. All animal pro- cultured and wild puffer sh Fugu niphobles. Bull. Jpn. Soc. Sci. Fish. 48, 253. Miyazawa, K., Noguchi, T., 2001. Distribution and origin of tetrodotoxin. J. Toxicol. cedures comply with the Japanese Government Animal Protection Toxin Rev. 20, 11e33. and Management Law (No. 105) and Japanese Government Notifi- Mosher, H.S., Fuhrman, G.J., Fuhrman, F.A., Fischer, H.G., 1964. Tarichatoxin-tetro- e cation on Feeding and Safekeeping of Animals (No. 6). dotoxin, a potent neurotoxin. Science 144, 1100 1110. Narahashi, T., 2001. Pharmacology of tetrodotoxin. J. Toxicol. Toxin Rev. 20, 67e84. Noguchi, T., Arakawa, O., 2008. Tetrodotoxin e distribution and accumulation in Transparency document aquatic organisms, and cases of human intoxication. Mar. Drugs 6, 220e242. Noguchi, T., Arakawa, O., Takatani, T., 2006a. TTX accumulation in pufferfish. Comp. Biochem. Physiol. D 1, 145e152. Transparency document related to this article can be found Noguchi, T., Arakawa, O., Takatani, T., 2006b. Toxicity of pufferfish Takifugu rubripes online at http://dx.doi.org/10.1016/j.toxicon.2015.10.009. cultured in netcages at sea or aquaria on land. Comp. Biochem. Physiol. D 1, 153e157. Noguchi, T., Hwang, D.F., Arakawa, O., Sugita, H., Deguchi, Y., Shida, Y., Appendix A. Supplementary data Hashimoto, K., 1987. Vibrio alginolyticus, a tetrodotoxin-producing bacterium, in the intestine of the fish Fugu vermicularis vermicularis. Mar. Biol. 94, 625e630. Noguchi, T., Maruyama, J., Ueda, Y., Hashimoto, K., Harada, T., 1981. Occurrence of Supplementary data related to this article can be found at http:// tetrodotoxin in the Japanese ivory shell Babylonia japonica. Bull. Jpn. Soc. Sci. dx.doi.org/10.1016/j.toxicon.2015.10.009. Fish. 47, 901e913. Noguchi, T., Uzu, A., Koyama, K., Hashimoto, K., 1983. Occurrence of tetrodotoxin as the major toxin in the xanthid crab Atergatis floridus. Bull. Jpn. Soc. Sci. Fish. 49, References 1887e1892. Okita, K., Yamazaki, H., Sakiyama, K., Yamane, H., Niina, S., Takatani, T., Arakawa, O., Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman, D.J., Sakakura, Y., 2013. Puffer smells tetrodotoxin. Ichthyol. Res. 60, 386e389. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database Palumbi, S., Martin, A., Romano, S., McMillan, W.O., Stice, L., Grabowski, G., 1991. The search programs. Nucleic Acids Res. 25, 3389e3402. Simple Fool's Guide to PCR, Version 2. Department of Zoology and Kewalo Asakawa, M., Ito, K., Kajihara, H., 2013. Highly toxic ribbon worm Cephalothrix Marine Laboratory, University of Hawaii, Honolulu. simula containing tetrodotoxin in Hiroshima Bay, Hiroshima Prefecture, Japan. Ritson-Williams, R., Yotsu-Yamashita, M., Paul, V.J., 2006. Ecological functions of Toxin 5, 376e395. tetrodotoxin in a deadly polyclad flatworm. Proc. Natl. Acad. Sci. U. S. A. 103, Cardall, B.L., Brodie, E.D., Hanifin, C.T., 2004. Secretion and regeneration of tetro- 3176e3179. dotoxin in the rough-skin newt (Taricha granulosa). Toxicon 44, 933e938. Saito, T., Kageyu, K., Goto, H., Murakami, K., Noguchi, T., 1997. Tetrodotoxin attracts Colquhon, D., Henderson, R., Ritchie, J.M., 1972. The binding of labeled tetrodotoxin pufferfish (‘torafugu’, Takifugu rubripes). Toxicon 35, 489. to non-myelineated nerve fibres. J. Physiol. 227, 95e126. Saito, T., Maruyama, J., Kanoh, S., Jeon, J.-K., Noguchi, T., Harada, T., Murata, O., Do, H.K., Kogure, K., Simidu, U., 1990. Identification of deep-sea-sediment bacteria Hashimoto, K., 1984. Toxicity of the cultured pufferfish Fugu rubripes rubripes which produce tetrodotoxin. Appl. Environ. Microbiol. 56, 1162e1163. along with their resistibility against tetrodotoxin. Bull. Jpn. Soc. Sci. Fish. 50, Elliott, S.A., Kats, L.B., Breeding, J.A., 1993. The use of conspecific chemical cues for 1573e1575. cannibal avoidance in California newts (Taricha torosa). Ethology 95, 186e192. Sezaki, K., Begum, R.A., Wongrat, P., Srivastava, M.P., SriKantha, S., Kikuchi, K., Food Safety Commission of Japan, 2005. Food Safety Risk Assessment Concerning Ishihara, H., Tanaka, S., Taniuchi, T., Watabe, S., 1999. Molecular phylogeny of “Liver of Takifugu rubripes Cultivated by the Method Proposed by Saga Prefec- Asian freshwater and marine stingrays based on the DNA nucleotide and ture and Ureshinocho in the Prefecture under the Law on Special Zones for deduced amino acid sequence of the cytochrome b gene. Fish. Sci. 65, 563e570. Structural Reform (Law No. 189, 2002)” (Risk Assessment Reports e Natural Sheumack, D.D., Howden, M.E., Spence, I., Quinn, R.J., 1978. Maculotoxin: a neuro- toxins/mycotoxins). toxin from the venom glands of the octopus Hapalochlaena maculosa identified Hanifin, C.T., Brodie III, E.D., Brodie Jr., E.D., 2002. Tetrodotoxin levels of the rough- as tetrodotoxin. Science 188e189. skin newt, Taricha granulosa, increase in long-term captivity. Toxicon 40, Simidu, U., Noguchi, T., Hwang, D.F., Shida, Y., Hashimoto, K., 1987. Marine bacteria 1149e1153. which produce tetrodotoxin. Appl. Environ. Microbiol. 53, 1714e1715. Hanifin, C.T., Brodie III, E.D., Brodie Jr., E.D., 2003. Tetrodotoxin levels in eggs of the Shinno, M., Nabae, Y., Higashi, T., Oka, Y., Takechi, T., Ogasawara, M., Oseto, M., rough-skin newt, Taricha granulosa, are correlated with female toxicity. J. Chem. Inouye, H., 2007. Determination of tetrodotoxin by liquid chromatography with Ecol. 29, 1729e1739. tandem mass spectrometry in the case of puffer fish poisoning. Annu. Rep. Honda, S., Arakawa, O., Takatani, T., Tachibana, K., Yagi, M., Tanigawa, A., Noguchi, T., Ehime Pref. Inst. Publ. Health Environ. Sci. 10, 14e17. 2005. Toxification of cultured puffer fish Takifugu rubripes by feeding on Stokes, A.N., Ducey, P.K., Neuman-Lee, L., Hanifin, C.T., French, S.S., Pfrender, M.E., tetrodotoxin-containing diet. Nippon. Suisan Gakkaishi 71, 815e820. Brodie III, E.D., Brodie Jr., E.D., 2014. Confirmation and distribution of tetrodo- Hwang, P.A., Noguchi, T., Hwang, D.F., 2004. Neurotoxin tetrodotoxin as attractant toxin for the first time in terrestrial invertebrates: two terrestrial flatworm for toxic snails. Fish. Sci. 70, 1106e1112. species (Bipalium adventitium and Bipalium kewense). PLoS One 9, e100718. Ikeda, K., Murakami, Y., Emoto, Y., Ngy, L., Taniyama, S., Yagi, M., Takatani, T., Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: molecular Arakawa, O., 2009. Transfer profile of intramuscularly administrated tetrodo- evolutionary genetics analysis version 6.0. Mol. Biol. Evol. 30, 2725e2729. toxin to non-toxic cultured specimens of the pufferfish Takifugu rubripes.Tox- Tani, I., 1945. Toxicological Studies on Japanese Puffers. Teikoku Tosho, Tokyo, p. 103. icon 53, 99e103. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensi- Ikeda, K., Emoto, Y., Tatsuno, R., Wang, J.J., Ngy, L., Taniyama, S., Takatani, T., tivity of progressive multiple sequence alignment through sequence weighting, Arakawa, O., 2010. Maturation-associated changes in toxicity of the pufferfish position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, Takifugu poecilonotus. Toxicon 55, 289e297. 4673e4680. Itoi, S., Yoshikawa, S., Tatsuno, R., Suzuki, M., Asahina, K., Yamamoto, S., Tsuruda, K., Arakawa, O., Kawatsu, K., Hamano, Y., Takatani, T., Noguchi, T., 2002. Takanashi, S., Takatani, T., Arakawa, O., Sakakura, Y., Sugita, H., 2012. Difference Secretory glands of tetrodotoxin in the skin of the Japanese newt Cynops pyr- in the localization of tetrodotoxin between the female and male pufferfish rhogaster. Toxicon 40, 131e136. Takifugu niphobles, during spawning. Toxicon 60, 1000e1004. Williams, B.L., 2010. Behavioral and chemical ecology of marine organisms with Itoi, S., Yoshikawa, S., Asahina, K., Suzuki, M., Ishizuka, K., Takimoto, N., Mitsuoka, R., respect to tetrodotoxin. Mar. Drugs 8, 381e398. Yokoyama, N., Detake, A., Takayanagi, C., Eguchi, M., Tatsuno, R., Kawane, M., Wu, Z., Xie, L., Xia, G., Zhang, J., Nie, Y., Hu, J., Wang, S., Zhang, R., 2005. A new Kokubo, S., Takanashi, S., Miura, A., Suitoh, K., Takatani, T., Arakawa, O., tetrodotoxin-producing actinomycete, Nocardiopsis dassonvillei, isolated from Sakakura, Y., Sugita, H., 2014. Larval pufferfish protected by maternal tetrodo- the ovaries of puffer fish Fugu rubripes. Toxicon 45, 851e859. toxin. Toxicon 78, 35e40. Wang, X.J., Yu, R.C., Luo, X., Zhou, M.J., Lin, X.T., 2008. Toxin-screening and identi- Kats, L.B., Elliott, S.A., Currens, J., 1992. Intraspecific oophagy in stream-breeding fication of bacteria isolated from highly toxic marine gastropod Nassarius California newts Taricha torosa. Herpetol. Rev. 23, 7e8. semiplicatus. Toxicon 52, 55e61. Kim, Y.H., Brown, G.B., Mosher, H.S., Fuhrman, F.A., 1975. Tetrodotoxin: occurrence Yamamori, K., Kono, M., Furukawa, K., Matsui, T., 2004. The toxification of juvenile in atelopid frogs of costa rica. Science 189, 151e152. cultured kusafugu Takifugu niphobles by oral administration of crystalline Kodama, M., Ogata, T., Sato, S., 1985. External secretion of tetrodotoxin from puffer tetrodotoxin. J. Food Hyg. Soc. Jpn. 45, 73e75. fishes stimulated by electric shock. Mar. Biol. 87, 199e202. Yang, G., Xu, J., Liang, S., Ren, D., Yan, X., Bao, B., 2010. A novel TTX-producing Maruyama, J., Noguchi, T., Jeon, J.K., Harada, T., Hashimoto, K., 1984. Occurrence of Aeromonas isolated from the ovary of Takifugu obscurus. Toxicon 56, 324e329.